Noise tolerant conductance-based logic gate and methods of operation and manufacturing thereof

Abstract

A logic gate, an adder and methods of operating and manufacturing the same. In one embodiment, the logic gate includes: (1) a summer, having at least two single-bit inputs and a noise-suppression input with corresponding conductances representing discrete weights, that generates a weighted sum of input binary digits presented at the at least two single-bit inputs and the noise-suppression input and (2) a quantizer, coupled to the summer, that generates an output binary digit at a binary output thereof that is a function of the weighted sum, the noise-suppression input increasing a noise tolerance of the logic gate.

What is claimed is:  
     
       1. A logic gate, comprising: 
       a summer, having at least two single-bit inputs and a noise-suppression input with corresponding conductances representing discrete weights, that generates a weighted sum of input binary digits presented at said at least two single-bit inputs and said noise-suppression input; and  
       a quantizer, coupled to said summer, that generates an output binary digit at a binary output thereof that is a function of said weighted sum, said noise-suppression input increasing a noise tolerance of said logic gate.  
     
     
       2. The gate as recited in  claim 1  wherein said summer comprises a noise-suppression logic circuit that develops said noise-suppression input. 
     
     
       3. The gate as recited in  claim 2  wherein some of said input binary digits presented at said at least two single-bit inputs are also presented at said noise-suppression logic circuit. 
     
     
       4. The gate as recited in  claim 2  wherein all of said input binary digits presented at said at least two single-bit inputs are also presented at said noise-suppression logic circuit. 
     
     
       5. The gate as recited in  claim 1  wherein said discrete weights are integer multiples of a predetermined number. 
     
     
       6. The gate as recited in  claim 1  wherein each of said at least two single-bit inputs and said noise-suppression input comprises: 
       a voltage source; and  
       a transistor, coupled to said voltage source and having a selectable intrinsic conductance, that couples said voltage source to said summer as a function of a corresponding particular input binary digit.  
     
     
       7. The gate as recited in  claim 6  wherein said transistor has a size that determines said selectable intrinsic conductance. 
     
     
       8. The gate as recited in  claim 6  wherein said transistor is selected from the group consisting of: 
       a p-channel metal oxide semiconductor (MOS) transistor, and  
       an n-channel MOS transistor.  
     
     
       9. The gate as recited in  claim 6  wherein a greater one of said discrete weights is associated with a one of said at least two single-bit inputs having a p-channel MOS transistor. 
     
     
       10. The gate as recited in  claim 1  further comprising a threshold input that provides a threshold number to said quantizer, said output binary digit being a function of a relationship between said weighted sum and said threshold number. 
     
     
       11. The gate as recited in  claim 10  wherein said threshold number is determined by relative sizes of at least two transistors forming said quantizer. 
     
     
       12. The gate as recited in  claim 1  wherein said quantizer comprises an inverter. 
     
     
       13. A method of performing a logic operation with respect to input binary digits, comprising: 
       employing conductances representing discrete weights and corresponding to at least two single-bit inputs and a noise-suppression input to generate a weighted sum of said input binary digits presented at said at least two single-bit inputs and said noise-suppression input; and  
       generating an output binary digit that is a function of said weighted sum, said noise-suppression input increasing a noise tolerance of said generating.  
     
     
       14. The method as recited in  claim 13  further comprising developing said noise-suppression input with a noise-suppression logic circuit. 
     
     
       15. The method as recited in  claim 13  wherein said noise-suppression input comprises a combination of some of said input binary digits. 
     
     
       16. The method as recited in  claim 13  wherein said noise-suppression input comprises a combination of all of said input binary digits. 
     
     
       17. The method as recited in  claim 13  wherein said discrete weights are integer multiples of a predetermined number. 
     
     
       18. The method as recited in  claim 13  wherein said employing comprises: 
       deriving a voltage from a voltage source; and  
       passing said voltage through a transistor having a selectable intrinsic conductance as a function of a corresponding particular input binary digit.  
     
     
       19. The method as recited in  claim 18  wherein said transistor has a size that determines said selectable intrinsic conductance. 
     
     
       20. The method as recited in  claim 18  wherein said transistor is selected from the group consisting of: 
       a p-channel metal oxide semiconductor (MOS) transistor, and  
       an n-channel MOS transistor.  
     
     
       21. The method as recited in  claim 18  wherein a greater one of said discrete weights is associated with a one of said at least two single-bit inputs having a p-channel MOS transistor. 
     
     
       22. The method as recited in  claim 13  wherein said generating further comprises generating said output binary digit as a function of a relationship between said weighted sum and a threshold number. 
     
     
       23. The method as recited in  claim 13  wherein said generating comprises inverting said weighted sum. 
     
     
       24. A method of manufacturing a logic gate, comprising: 
       providing a summer having at least two single-bit inputs and a noise-suppression input with corresponding conductances representing discrete weights, said summer capable of generating a weighted sum of input binary digits presented at said at least two single-bit inputs and said noise-suppression input; and  
       coupling a quantizer to said summer, said quantizer capable of generating an output binary digit at a binary output thereof that is a function of said weighted sum, said noise-suppression input increasing a noise tolerance of said logic gate.  
     
     
       25. The method as recited in  claim 24  wherein said summer comprises a noise-suppression logic circuit that develops said noise-suppression input. 
     
     
       26. The method as recited in  claim 24  wherein some of said input binary digits presented at said at least two single-bit inputs are also presented at said noise-suppression logic circuit. 
     
     
       27. The method as recited in  claim 24  wherein all of said input binary digits presented at said at least two single-bit inputs are also presented at said noise-suppression logic circuit. 
     
     
       28. The method as recited in  claim 24  wherein said discrete weights are integer multiples of a predetermined number. 
     
     
       29. The method as recited in  claim 24  wherein each of said at least two single-bit inputs and said noise-suppression input comprises: 
       a voltage source; and  
       a transistor, coupled to said voltage source and having a selectable intrinsic conductance, that couples said voltage source to said summer as a function of a corresponding particular input binary digit.  
     
     
       30. The method as recited in  claim 29  wherein said transistor has a size that determines said selectable intrinsic conductance. 
     
     
       31. The method as recited in  claim 29  wherein said transistor is selected from the group consisting of: 
       a p-channel metal oxide semiconductor (MOS) transistor, and  
       an n-channel MOS transistor.  
     
     
       32. The method as recited in  claim 29  wherein a greater one of said discrete weights is associated with a one of said at least two single-bit inputs having a p-channel MOS transistor. 
     
     
       33. The method as recited in  claim 24  further comprising a threshold input that provides a threshold number to said quantizer, said output binary digit being a function of a relationship between said weighted sum and said threshold number. 
     
     
       34. The method as recited in  claim 33  further comprising encoding said threshold number in the relative sizes of at least two transistors forming said quantizer. 
     
     
       35. The method as recited in  claim 24  wherein said quantizer comprises an inverter. 
     
     
       36. A logic gate, comprising: 
       a summer, having at least two single-bit inputs and an anti-floating input with corresponding conductances representing discrete weights, that generates a weighted sum of input binary digits presented at said at least two single-bit inputs and said anti-floating input; and  
       a quantizer, coupled to said summer, that generates an output binary digit at a binary output thereof that is a function of said weighted sum, said anti-floating input preventing said weighted sum from being in an indefinite state.  
     
     
       37. The gate as recited in  claim 36  wherein said summer comprises an anti-floating logic circuit that develops said anti-floating input. 
     
     
       38. The gate as recited in  claim 37  wherein said anti-floating logic circuit implements an AND logic function. 
     
     
       39. The gate as recited in  claim 37  wherein some of said input binary digits presented at said at least two single-bit inputs are also presented at said anti-floating logic circuit. 
     
     
       40. The gate as recited in  claim 37  wherein all of said input binary digits presented at said at least two single-bit inputs are also presented at said anti-floating logic circuit. 
     
     
       41. The gate as recited in  claim 36  wherein said discrete weights are integer multiples of a predetermined number. 
     
     
       42. The gate as recited in  claim 36  wherein each of said at least two single-bit inputs and said anti-floating input comprises: 
       a voltage source; and  
       a transistor, coupled to said voltage source and having a selectable intrinsic conductance, that couples said voltage source to said summer as a function of a corresponding particular input binary digit.  
     
     
       43. The gate as recited in  claim 42  wherein said transistor has a size that determines said selectable intrinsic conductance. 
     
     
       44. The gate as recited in  claim 42  wherein said transistor is selected from the group consisting of: 
       a p-channel metal oxide semiconductor (MOS) transistor, and  
       an n-channel MOS transistor.  
     
     
       45. The gate as recited in  claim 42  wherein a greater one of said discrete weights is associated with a one of said at least two single-bit inputs having a p-channel MOS transistor. 
     
     
       46. The gate as recited in  claim 36  further comprising a threshold input that provides a threshold number to said quantizer, said output binary digit being a function of a relationship between said weighted sum and said threshold number. 
     
     
       47. The gate as recited in  claim 46  wherein said threshold number is determined by relative sizes of at least two transistors forming said quantizer. 
     
     
       48. The gate as recited in  claim 36  wherein said quantizer comprises an inverter. 
     
     
       49. A method of performing a logic operation with respect to input binary digits, comprising: 
       employing conductances representing discrete weights and corresponding to at least two single-bit inputs and an anti-floating input to generate a weighted sum of said input binary digits presented at said at least two single-bit inputs and said anti-floating input; and  
       generating an output binary digit that is a function of said weighted sum, said anti-floating input preventing said weighted sum from being in an indefinite state.  
     
     
       50. The method as recited in  claim 49  further comprising developing said anti-floating input with an anti-floating logic circuit. 
     
     
       51. The method as recited in  claim 49  wherein said anti-floating logic circuit implements an AND logic function. 
     
     
       52. The method as recited in  claim 49  wherein said anti-floating input comprises a combination of some of said input binary digits. 
     
     
       53. The method as recited in  claim 49  wherein said anti-floating input comprises a combination of all of said input binary digits. 
     
     
       54. The method as recited in  claim 49  wherein said discrete weights are integer multiples of a predetermined number. 
     
     
       55. The method as recited in  claim 49  wherein said employing comprises: 
       deriving a voltage from a voltage source; and  
       passing said voltage through a transistor having a selectable intrinsic conductance as a function of a corresponding particular input binary digit.  
     
     
       56. The method as recited in  claim 55  wherein said transistor has a size that determines said selectable intrinsic conductance. 
     
     
       57. The method as recited in  claim 55  wherein said transistor is selected from the group consisting of: 
       a p-channel metal oxide semiconductor (MOS) transistor, and  
       an n-channel MOS transistor.  
     
     
       58. The method as recited in  claim 55  wherein a greater one of said discrete weights is associated with a one of said at least two single-bit inputs having a p-channel MOS transistor. 
     
     
       59. The method as recited in  claim 49  wherein said generating further comprises generating said output binary digit as a function of a relationship between said weighted sum and a threshold number. 
     
     
       60. The method as recited in  claim 49  wherein said generating comprises inverting said weighted sum. 
     
     
       61. A method of manufacturing a logic gate, comprising: 
       providing a summer having at least two single-bit inputs and an anti-floating input with corresponding conductances representing discrete weights, said summer capable of generating a weighted sum of input binary digits presented at said at least two single-bit inputs and said anti-floating input; and  
       coupling a quantizer to said summer, said quantizer capable of generating an output binary digit at a binary output thereof that is a function of said weighted sum, said anti-floating input preventing said weighted sum from being in an indefinite state.  
     
     
       62. The method as recited in  claim 61  wherein said summer comprises an anti-floating logic circuit that develops said anti-floating input. 
     
     
       63. The method as recited in  claim 61  wherein said anti-floating logic circuit implements an AND logic function. 
     
     
       64. The method as recited in  claim 61  wherein some of said input binary digits presented at said at least two single-bit inputs are also presented at said anti-floating logic circuit. 
     
     
       65. The method as recited in  claim 61  wherein all of said input binary digits presented at said at least two single-bit inputs are also presented at said anti-floating logic circuit. 
     
     
       66. The method as recited in  claim 61  wherein said discrete weights are integer multiples of a predetermined number. 
     
     
       67. The method as recited in  claim 61  wherein each of said at least two single-bit inputs and said anti-floating input comprises: 
       a voltage source; and  
       a transistor, coupled to said voltage source and having a selectable intrinsic conductance, that couples said voltage source to said summer as a function of a corresponding particular input binary digit.  
     
     
       68. The method as recited in  claim 67  wherein said transistor has a size that determines said selectable intrinsic conductance. 
     
     
       69. The method as recited in  claim 67  wherein said transistor is selected from the group consisting of: 
       a p-channel metal oxide semiconductor (MOS) transistor, and  
       an n-channel MOS transistor.  
     
     
       70. The method as recited in  claim 67  wherein a greater one of said discrete weights is associated with a one of said at least two single-bit inputs having a p-channel MOS transistor. 
     
     
       71. The method as recited in  claim 61  wherein said quantizer comprises a threshold input that provides a threshold number to said quantizer, said output binary digit being a function of a relationship between said weighted sum and said threshold number. 
     
     
       72. The method as recited in  claim 71  further comprising encoding said threshold number in the relative sizes of at least two transistors forming said quantizer. 
     
     
       73. The method as recited in  claim 61  wherein said quantizer comprises an inverter. 
     
     
       74. An adder, comprising: 
       a plurality of logic gates for deriving a carry-out bit, each of said plurality of logic gates including:  
       a summer, having at least two single-bit inputs a noise-suppression input and an anti-floating input with corresponding conductances representing discrete weights, that generates a weighted sum of input binary digits presented at said at least two single-bit inputs, said noise-suppression input and said anti-floating input, and  
       a quantizer, coupled to said summer, that generates an output binary digit at a binary output thereof that is a function of said weighted sum, said noise-suppression input increasing a noise tolerance of said logic gate, said anti-floating input preventing said weighted sum from being in an indefinite state.  
     
     
       75. The adder as recited in  claim 74  wherein at least some of said plurality of logic gates comprise first, second, third and fourth single-bit inputs having discrete weights of 1, 1, 2 and 2, respectively. 
     
     
       76. The adder as recited in  claim 74  further comprising a circuit for deriving at least one group-carry-generate bit. 
     
     
       77. A method of adding, comprising: 
       deriving a carry-out bit with a plurality of logic gates, each of said plurality of logic gates including:  
       a summer, having at least two single-bit inputs, a noise-suppression input and an anti-floating input with corresponding conductances representing discrete weights, that generates a weighted sum of input binary digits presented at said at least two single-bit inputs, said noise-suppression input and said anti-floating input, and  
       a quantizer, coupled to said summer, that generates an output binary digit at a binary output thereof that is a function of said weighted sum, said noise-suppression input increasing a noise tolerance of said logic gate, said anti-floating input preventing said weighted sum from being in an indefinite state.  
     
     
       78. The method as recited in  claim 77  wherein at least some of said plurality of logic gates comprise first, second, third and fourth binary inputs having discrete weights of 1, 1, 2 and 2, respectively. 
     
     
       79. A method of manufacturing an adder, comprising: 
       assembling a plurality of logic gates to derive a carry-out bit, each of said plurality of logic gates including:  
       a summer, having at least two single-bit inputs, a noise-suppression input and an anti-floating input with corresponding conductances representing discrete weights, that generates a weighted sum of input binary digits presented at said at least two single-bit inputs, said noise-suppression input and said anti-floating input, and  
       a quantizer, coupled to said summer, that generates an output binary digit at a binary output thereof that is a function of said weighted sum, said noise-suppression input increasing a noise tolerance of said logic gate, said anti-floating input preventing said weighted sum from being in an indefinite state.  
     
     
       80. The method as recited in  claim 79  wherein at least some of said plurality of logic gates comprise first, second, third and fourth binary inputs having discrete weights of 1, 1, 2 and 2, respectively.